Data science Project (University ranking)¶

Import libraries
In [1]:
import pandas as pd
import numpy as np
import seaborn as sns
import matplotlib.pyplot as plt
import plotly.express as px
import pycountry
import pycountry_convert as pc
from mpl_toolkits.basemap import Basemap
from geopy.geocoders import Nominatim
from random import randint

import warnings
warnings.filterwarnings('ignore')
Import Dataset
In [2]:
# import file
data = pd.read_csv ('uni_dataset.csv', encoding= 'unicode_escape')
data
Out[2]:
World Rank Institution Location National Rank Quality of Education Alumni Employment Quality of Faculty Research Output Quality Publications Influence Citations Score
0 1 Harvard University USA 1 2 1 1 1 1 1 1 100.0
1 2 Stanford University USA 2 10 3 2 10 4 3 2 96.7
2 3 Massachusetts Institute of Technology USA 3 3 11 3 30 15 2 6 95.1
3 4 University of Cambridge United Kingdom 1 5 19 6 12 8 6 19 94.0
4 5 University of Oxford United Kingdom 2 9 25 10 9 5 7 4 93.2
... ... ... ... ... ... ... ... ... ... ... ... ...
995 996 Aga Khan University Pakistan 3 - > 1000 - > 1000 > 1000 464 673 69.8
996 997 University of Calcutta India 17 353 716 296 798 966 > 1000 > 1000 69.8
997 998 K?chi University Japan 56 - > 1000 - > 1000 > 1000 811 673 69.8
998 999 Soonchunhyang University South Korea 35 - > 1000 - 881 > 1000 > 1000 898 69.8
999 1000 Capital Normal University China 108 - 869 - 923 904 889 > 1000 69.8

1000 rows × 12 columns

In [3]:
# creat dataframe
df = pd.DataFrame (data)
df
Out[3]:
World Rank Institution Location National Rank Quality of Education Alumni Employment Quality of Faculty Research Output Quality Publications Influence Citations Score
0 1 Harvard University USA 1 2 1 1 1 1 1 1 100.0
1 2 Stanford University USA 2 10 3 2 10 4 3 2 96.7
2 3 Massachusetts Institute of Technology USA 3 3 11 3 30 15 2 6 95.1
3 4 University of Cambridge United Kingdom 1 5 19 6 12 8 6 19 94.0
4 5 University of Oxford United Kingdom 2 9 25 10 9 5 7 4 93.2
... ... ... ... ... ... ... ... ... ... ... ... ...
995 996 Aga Khan University Pakistan 3 - > 1000 - > 1000 > 1000 464 673 69.8
996 997 University of Calcutta India 17 353 716 296 798 966 > 1000 > 1000 69.8
997 998 K?chi University Japan 56 - > 1000 - > 1000 > 1000 811 673 69.8
998 999 Soonchunhyang University South Korea 35 - > 1000 - 881 > 1000 > 1000 898 69.8
999 1000 Capital Normal University China 108 - 869 - 923 904 889 > 1000 69.8

1000 rows × 12 columns

Preprocessing
In [4]:
df.columns.tolist()
Out[4]:
['World Rank',
 'Institution',
 'Location',
 'National Rank',
 'Quality\xa0of Education',
 'Alumni Employment',
 'Quality\xa0of Faculty',
 'Research Output',
 'Quality Publications',
 'Influence',
 'Citations',
 'Score']
In [5]:
new_columns_names = ["Quality of Education", "Quality of Faculty"]
dic = {"Quality\xa0of Education": new_columns_names[0], "Quality\xa0of Faculty": new_columns_names[1]}
df = df.rename(columns=dic)
In [6]:
df.columns.tolist()
Out[6]:
['World Rank',
 'Institution',
 'Location',
 'National Rank',
 'Quality of Education',
 'Alumni Employment',
 'Quality of Faculty',
 'Research Output',
 'Quality Publications',
 'Influence',
 'Citations',
 'Score']
In [7]:
# information about dataframe
df.describe(include="all")
Out[7]:
World Rank Institution Location National Rank Quality of Education Alumni Employment Quality of Faculty Research Output Quality Publications Influence Citations Score
count 1000.000000 1000 1000 1000.00000 1000 1000 1000 1000 1000 1000 1000 1000.000000
unique NaN 1000 61 NaN 404 492 221 905 892 830 75 NaN
top NaN Harvard University USA NaN - > 1000 - > 1000 > 1000 > 1000 898 NaN
freq NaN 1 213 NaN 597 509 731 77 51 171 129 NaN
mean 500.500000 NaN NaN 40.02300 NaN NaN NaN NaN NaN NaN NaN 75.034300
std 288.819436 NaN NaN 49.22871 NaN NaN NaN NaN NaN NaN NaN 4.670858
min 1.000000 NaN NaN 1.00000 NaN NaN NaN NaN NaN NaN NaN 69.800000
25% 250.750000 NaN NaN 6.00000 NaN NaN NaN NaN NaN NaN NaN 71.500000
50% 500.500000 NaN NaN 21.00000 NaN NaN NaN NaN NaN NaN NaN 73.750000
75% 750.250000 NaN NaN 51.00000 NaN NaN NaN NaN NaN NaN NaN 77.300000
max 1000.000000 NaN NaN 213.00000 NaN NaN NaN NaN NaN NaN NaN 100.000000
In [8]:
#check for "-" values
for col in df.columns :
    print(col,' : ','-' in set(df[col]))

World Rank  :  False
Institution  :  False
Location  :  False
National Rank  :  False
Quality of Education  :  True
Alumni Employment  :  False
Quality of Faculty  :  True
Research Output  :  False
Quality Publications  :  False
Influence  :  False
Citations  :  False
Score  :  False
In [9]:
#a closer look at the values - 
QE = df[df["Quality of Education"]=="-"]
print("Max Score in dataframe : ",QE.Score.max())
print("Min Score in dataframe : ",QE.Score.min())
print("Max World Rank in dataframe : ",QE["World Rank"].max())
print("Min World Rank in dataframe : ",QE["World Rank"].min())
QE

Max Score in dataframe :  87.7
Min Score in dataframe :  69.8
Max World Rank in dataframe :  1000
Min World Rank in dataframe :  22
Out[9]:
World Rank Institution Location National Rank Quality of Education Alumni Employment Quality of Faculty Research Output Quality Publications Influence Citations Score
21 22 University of California, San Francisco USA 17 - > 1000 13 45 23 8 17 87.7
59 60 Seoul National University South Korea 1 - 20 - 19 39 146 129 83.7
79 80 Emory University USA 47 - 197 164 101 66 55 63 82.5
88 89 University of Texas MD Anderson Cancer Center USA 49 - > 1000 91 158 104 70 66 82.0
93 94 Baylor College of Medicine USA 51 - > 1000 141 161 118 38 53 81.8
... ... ... ... ... ... ... ... ... ... ... ... ...
994 995 University of Macau Macau 1 - 512 - 883 816 > 1000 > 1000 69.8
995 996 Aga Khan University Pakistan 3 - > 1000 - > 1000 > 1000 464 673 69.8
997 998 K?chi University Japan 56 - > 1000 - > 1000 > 1000 811 673 69.8
998 999 Soonchunhyang University South Korea 35 - > 1000 - 881 > 1000 > 1000 898 69.8
999 1000 Capital Normal University China 108 - 869 - 923 904 889 > 1000 69.8

597 rows × 12 columns

In [10]:
dfQualityF = df[df["Quality of Faculty"]=="-"]
print("Max Score in dataframe : ",dfQualityF.Score.max())
print("Min Score in dataframe : ",dfQualityF.Score.min())
print("Max World Rank in dataframe : ",dfQualityF["World Rank"].max())
print("Min World Rank in dataframe : ",dfQualityF["World Rank"].min())
dfQualityF

Max Score in dataframe :  83.7
Min Score in dataframe :  69.8
Max World Rank in dataframe :  1000
Min World Rank in dataframe :  60
Out[10]:
World Rank Institution Location National Rank Quality of Education Alumni Employment Quality of Faculty Research Output Quality Publications Influence Citations Score
59 60 Seoul National University South Korea 1 - 20 - 19 39 146 129 83.7
72 73 University of Pittsburgh USA 43 187 233 - 34 25 47 18 82.9
91 92 Peking University China 1 340 67 - 23 38 88 167 81.9
97 98 Tsinghua University China 2 586 109 - 22 41 96 108 81.6
102 103 National University of Singapore Singapore 1 - 148 - 38 32 94 85 81.4
... ... ... ... ... ... ... ... ... ... ... ... ...
994 995 University of Macau Macau 1 - 512 - 883 816 > 1000 > 1000 69.8
995 996 Aga Khan University Pakistan 3 - > 1000 - > 1000 > 1000 464 673 69.8
997 998 K?chi University Japan 56 - > 1000 - > 1000 > 1000 811 673 69.8
998 999 Soonchunhyang University South Korea 35 - > 1000 - 881 > 1000 > 1000 898 69.8
999 1000 Capital Normal University China 108 - 869 - 923 904 889 > 1000 69.8

731 rows × 12 columns

In [11]:
df[(df["Quality of Faculty"]=="-") &(df["Quality of Education"]=="-")]
Out[11]:
World Rank Institution Location National Rank Quality of Education Alumni Employment Quality of Faculty Research Output Quality Publications Influence Citations Score
59 60 Seoul National University South Korea 1 - 20 - 19 39 146 129 83.7
102 103 National University of Singapore Singapore 1 - 148 - 38 32 94 85 81.4
106 107 KU Leuven Belgium 2 - 429 - 48 47 84 66 81.2
126 127 VU University Amsterdam Netherlands 6 - 984 - 91 71 132 44 80.5
152 153 Icahn School of Medicine at Mount Sinai USA 65 - > 1000 - 202 145 59 101 79.6
... ... ... ... ... ... ... ... ... ... ... ... ...
994 995 University of Macau Macau 1 - 512 - 883 816 > 1000 > 1000 69.8
995 996 Aga Khan University Pakistan 3 - > 1000 - > 1000 > 1000 464 673 69.8
997 998 K?chi University Japan 56 - > 1000 - > 1000 > 1000 811 673 69.8
998 999 Soonchunhyang University South Korea 35 - > 1000 - 881 > 1000 > 1000 898 69.8
999 1000 Capital Normal University China 108 - 869 - 923 904 889 > 1000 69.8

551 rows × 12 columns

After various investigations, we come to the conclusion that the best way to fill in dashed values is to use one greater than the largest number in that column.

In [12]:
#Change > 1000 to 1001
df.replace("> 1000",'1001',inplace=True)
#Chnage - to max+1 
df["Quality of Education"].replace('-','667',inplace=True)
df["Quality of Faculty"].replace('-','304',inplace=True)
In [13]:
for col in df.drop(columns=['Institution','Location','Score']):#Drop the columns that we don't want to retype
    df[col]=df[col].astype(np.int64)
¶
In [14]:
df.shape
Out[14]:
(1000, 12)
In [15]:
#Remove missing values
df.dropna()
Out[15]:
World Rank Institution Location National Rank Quality of Education Alumni Employment Quality of Faculty Research Output Quality Publications Influence Citations Score
0 1 Harvard University USA 1 2 1 1 1 1 1 1 100.0
1 2 Stanford University USA 2 10 3 2 10 4 3 2 96.7
2 3 Massachusetts Institute of Technology USA 3 3 11 3 30 15 2 6 95.1
3 4 University of Cambridge United Kingdom 1 5 19 6 12 8 6 19 94.0
4 5 University of Oxford United Kingdom 2 9 25 10 9 5 7 4 93.2
... ... ... ... ... ... ... ... ... ... ... ... ...
995 996 Aga Khan University Pakistan 3 667 1001 304 1001 1001 464 673 69.8
996 997 University of Calcutta India 17 353 716 296 798 966 1001 1001 69.8
997 998 K?chi University Japan 56 667 1001 304 1001 1001 811 673 69.8
998 999 Soonchunhyang University South Korea 35 667 1001 304 881 1001 1001 898 69.8
999 1000 Capital Normal University China 108 667 869 304 923 904 889 1001 69.8

1000 rows × 12 columns

In [16]:
# Add name of continent 
df['Continent']=np.NaN
for i in range(len(df)):
    country_code = pc.country_name_to_country_alpha2(df.Location.iloc[i], cn_name_format="default")
    df['Continent'].iloc[i] = pc.country_alpha2_to_continent_code(country_code)
In [17]:
df.Continent.replace ( {'NA': 'North America','SA': 'South America', 'AS': 'Asia','OC': 'Australia','AF': 'Africa',
                      'EU': 'Europe'},inplace=True)
In [18]:
# Add iso name for plot
df['Country']= np.NaN

def alpha3(input):
    
    input_countries = [input]

    countries = {}
    for country in pycountry.countries:
        countries[country.name] = country.alpha_3
        for country in input_countries:
            codes = countries.get(country, input) 

    return(codes)  

for i in range(len(df)):
    df["Country"].iloc[i] = alpha3(df.Location.iloc[i])
In [19]:
df['Country'].replace({"South Korea":"KOR", "Taiwan":"TWN","Russia":"RUS",
                    "Czech Republic":"CZE","Iran":"IRN","Slovak Republic":"SVK","Macau":"MAC"},inplace=True)
In [21]:
# Add longitude and lattitude
df["Longitude"]=np.NaN
df["Latitude"]=np.NaN
geolocator = Nominatim(user_agent="koosha")
for i in range(len(df)):
    try:
        Loc = df.Location.iloc[i]+","+df.Institution.iloc[i]
        address=geolocator.geocode(Loc)
        df["Latitude"].iloc[i]=address.latitude
        df["Longitude"].iloc[i]=address.longitude
    
    except:
        Loc= df.Location.iloc[i]
        address=geolocator.geocode(Loc)
        df["Latitude"].iloc[i]=address.latitude
        df["Longitude"].iloc[i]=address.longitude
In [22]:
df.info()

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 1000 entries, 0 to 999
Data columns (total 16 columns):
 #   Column                Non-Null Count  Dtype  
---  ------                --------------  -----  
 0   World Rank            1000 non-null   int64  
 1   Institution           1000 non-null   object 
 2   Location              1000 non-null   object 
 3   National Rank         1000 non-null   int64  
 4   Quality of Education  1000 non-null   int64  
 5   Alumni Employment     1000 non-null   int64  
 6   Quality of Faculty    1000 non-null   int64  
 7   Research Output       1000 non-null   int64  
 8   Quality Publications  1000 non-null   int64  
 9   Influence             1000 non-null   int64  
 10  Citations             1000 non-null   int64  
 11  Score                 1000 non-null   float64
 12  Continent             1000 non-null   object 
 13  Country               1000 non-null   object 
 14  Longitude             1000 non-null   float64
 15  Latitude              1000 non-null   float64
dtypes: float64(3), int64(9), object(4)
memory usage: 125.1+ KB
In [23]:
df.isnull().sum()
Out[23]:
World Rank              0
Institution             0
Location                0
National Rank           0
Quality of Education    0
Alumni Employment       0
Quality of Faculty      0
Research Output         0
Quality Publications    0
Influence               0
Citations               0
Score                   0
Continent               0
Country                 0
Longitude               0
Latitude                0
dtype: int64
In [24]:
df.describe(include='all').T
Out[24]:
count unique top freq mean std min 25% 50% 75% max
World Rank 1000.0 NaN NaN NaN 500.5 288.819436 1.0 250.75 500.5 750.25 1000.0
Institution 1000 1000 Harvard University 1 NaN NaN NaN NaN NaN NaN NaN
Location 1000 61 USA 213 NaN NaN NaN NaN NaN NaN NaN
National Rank 1000.0 NaN NaN NaN 40.023 49.22871 1.0 6.0 21.0 51.0 213.0
Quality of Education 1000.0 NaN NaN NaN 528.532 208.781322 1.0 405.75 667.0 667.0 667.0
Alumni Employment 1000.0 NaN NaN NaN 726.528 343.686144 1.0 425.75 1001.0 1001.0 1001.0
Quality of Faculty 1000.0 NaN NaN NaN 261.18 84.233419 1.0 282.75 304.0 304.0 304.0
Research Output 1000.0 NaN NaN NaN 508.454 300.080433 1.0 250.75 500.5 754.25 1001.0
Quality Publications 1000.0 NaN NaN NaN 503.003 292.854863 1.0 250.75 500.5 750.25 1001.0
Influence 1000.0 NaN NaN NaN 529.156 322.419239 1.0 250.75 502.5 814.75 1001.0
Citations 1000.0 NaN NaN NaN 501.159 308.874517 1.0 240.0 460.0 673.0 1001.0
Score 1000.0 NaN NaN NaN 75.0343 4.670858 69.8 71.5 73.75 77.3 100.0
Continent 1000 6 Europe 388 NaN NaN NaN NaN NaN NaN NaN
Country 1000 61 USA 213 NaN NaN NaN NaN NaN NaN NaN
Longitude 1000.0 NaN NaN NaN 13.543153 80.867619 -157.818692 -72.255634 7.047121 104.055891 174.773733
Latitude 1000.0 NaN NaN NaN 35.959617 19.690072 -45.864979 33.832598 39.78373 46.603354 69.681907
In [25]:
df
Out[25]:
World Rank Institution Location National Rank Quality of Education Alumni Employment Quality of Faculty Research Output Quality Publications Influence Citations Score Continent Country Longitude Latitude
0 1 Harvard University USA 1 2 1 1 1 1 1 1 100.0 North America USA -71.126782 42.367909
1 2 Stanford University USA 2 10 3 2 10 4 3 2 96.7 North America USA -122.169365 37.431314
2 3 Massachusetts Institute of Technology USA 3 3 11 3 30 15 2 6 95.1 North America USA -71.096627 42.358253
3 4 University of Cambridge United Kingdom 1 5 19 6 12 8 6 19 94.0 Europe GBR 0.092005 52.210946
4 5 University of Oxford United Kingdom 2 9 25 10 9 5 7 4 93.2 Europe GBR -1.255668 51.758708
... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ...
995 996 Aga Khan University Pakistan 3 667 1001 304 1001 1001 464 673 69.8 Asia PAK 67.062022 24.923236
996 997 University of Calcutta India 17 353 716 296 798 966 1001 1001 69.8 Asia IND 88.344753 22.484733
997 998 K?chi University Japan 56 667 1001 304 1001 1001 811 673 69.8 Asia JPN 139.239418 36.574844
998 999 Soonchunhyang University South Korea 35 667 1001 304 881 1001 1001 898 69.8 Asia KOR 126.931634 36.769988
999 1000 Capital Normal University China 108 667 869 304 923 904 889 1001 69.8 Asia CHN 116.300055 39.928866

1000 rows × 16 columns

In [26]:
col = df.pop('Score')
df.insert(len(df.columns), 'Score', col)
Visualization
In [27]:
df
Out[27]:
World Rank Institution Location National Rank Quality of Education Alumni Employment Quality of Faculty Research Output Quality Publications Influence Citations Continent Country Longitude Latitude Score
0 1 Harvard University USA 1 2 1 1 1 1 1 1 North America USA -71.126782 42.367909 100.0
1 2 Stanford University USA 2 10 3 2 10 4 3 2 North America USA -122.169365 37.431314 96.7
2 3 Massachusetts Institute of Technology USA 3 3 11 3 30 15 2 6 North America USA -71.096627 42.358253 95.1
3 4 University of Cambridge United Kingdom 1 5 19 6 12 8 6 19 Europe GBR 0.092005 52.210946 94.0
4 5 University of Oxford United Kingdom 2 9 25 10 9 5 7 4 Europe GBR -1.255668 51.758708 93.2
... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ...
995 996 Aga Khan University Pakistan 3 667 1001 304 1001 1001 464 673 Asia PAK 67.062022 24.923236 69.8
996 997 University of Calcutta India 17 353 716 296 798 966 1001 1001 Asia IND 88.344753 22.484733 69.8
997 998 K?chi University Japan 56 667 1001 304 1001 1001 811 673 Asia JPN 139.239418 36.574844 69.8
998 999 Soonchunhyang University South Korea 35 667 1001 304 881 1001 1001 898 Asia KOR 126.931634 36.769988 69.8
999 1000 Capital Normal University China 108 667 869 304 923 904 889 1001 Asia CHN 116.300055 39.928866 69.8

1000 rows × 16 columns

In [28]:
features=df[['National Rank', 'Quality of Education', 'Alumni Employment',
       'Quality of Faculty', 'Research Output', 'Quality Publications',
       'Influence', 'Citations',  'Longitude','Latitude','Score']]
features
Out[28]:
National Rank Quality of Education Alumni Employment Quality of Faculty Research Output Quality Publications Influence Citations Longitude Latitude Score
0 1 2 1 1 1 1 1 1 -71.126782 42.367909 100.0
1 2 10 3 2 10 4 3 2 -122.169365 37.431314 96.7
2 3 3 11 3 30 15 2 6 -71.096627 42.358253 95.1
3 1 5 19 6 12 8 6 19 0.092005 52.210946 94.0
4 2 9 25 10 9 5 7 4 -1.255668 51.758708 93.2
... ... ... ... ... ... ... ... ... ... ... ...
995 3 667 1001 304 1001 1001 464 673 67.062022 24.923236 69.8
996 17 353 716 296 798 966 1001 1001 88.344753 22.484733 69.8
997 56 667 1001 304 1001 1001 811 673 139.239418 36.574844 69.8
998 35 667 1001 304 881 1001 1001 898 126.931634 36.769988 69.8
999 108 667 869 304 923 904 889 1001 116.300055 39.928866 69.8

1000 rows × 11 columns

In [29]:
#histplot with density estimate line
plt.figure(figsize=(22, 20))
for i, feature in enumerate(features):
    plt.subplot(4, 3, i+1)
    sns.histplot(df[feature], stat="density", kde=True)
    plt.title(f"Distribution of {feature}")
    
plt.tight_layout()

plt.show()
No description has been provided for this image
In [30]:
plt.figure(figsize=(11,9))
sns.heatmap(df.drop(columns=['Institution','Location','Country','Continent']).corr(),cmap='coolwarm',  annot=True, fmt=".2f")

plt.title('Correlation Matrix')
Out[30]:
Text(0.5, 1.0, 'Correlation Matrix')
No description has been provided for this image
In [31]:
#Scatter plot between all features and score "except Institution".
for column in df.drop(columns=['Score',"Institution",]).columns : 
    plt.figure(figsize=(15,10))
    plt.scatter(df[column],df.Score , alpha=.5 )
    plt.title(column+" & Score",backgroundcolor='orange',color='white',fontsize=30)
    plt.xticks(rotation=90)
    plt.xlabel(column,fontsize=20)
    plt.ylabel("Score",fontsize=20)
    plt.grid()
    plt.show()
No description has been provided for this image
No description has been provided for this image
No description has been provided for this image
No description has been provided for this image
No description has been provided for this image
No description has been provided for this image
No description has been provided for this image
No description has been provided for this image
No description has been provided for this image
No description has been provided for this image
No description has been provided for this image
No description has been provided for this image
No description has been provided for this image
No description has been provided for this image
In [32]:
plt.figure(figsize=(24,20))
for i in enumerate(features):
    plt.subplot(6,3,i[0]+1)
    sns.boxplot(x= i[1], data = df)
No description has been provided for this image
In [33]:
#make new dataframe of the first 100 World Rank 
df_100=df[df["World Rank"]<=100]
In [34]:
#Map plot
plt.figure(figsize=(30,30)) 
m = Basemap(projection='cyl',llcrnrlat=-50,urcrnrlat=90,llcrnrlon=-180,urcrnrlon=180,lat_ts=90,resolution='c') 
m.bluemarble() 
plt.scatter(df['Longitude'],df['Latitude'],1,marker='o',color='yellow',label='World Rank more than 100') 
plt.scatter(df_100['Longitude'],df_100['Latitude'],1,marker='o',color='Red',label='World Rank Top 100')
plt.title("Location of Universities",backgroundcolor='orange',color='white',fontsize=30)
plt.legend()
plt.show()
No description has been provided for this image
In [35]:
#Map plot
plt.figure(figsize=(20,15)) 
m = Basemap(projection='cyl',llcrnrlat=-50,urcrnrlat=90,llcrnrlon=-180,urcrnrlon=180,lat_ts=90,resolution='c')
m.bluemarble()
plt.scatter(df['Longitude'],df['Latitude'],1,marker='o',c=df["Score"],cmap="jet")
plt.title("Location of uni",backgroundcolor='orange',color='white',fontsize=30)
plt.colorbar(location ='bottom')
plt.show()
No description has been provided for this image
In [36]:
#Map plot
fig = px.scatter_mapbox(df, lat='Latitude', lon='Longitude', hover_name='Institution',color=df["Score"],
                        hover_data=["Continent","World Rank", "Score"],zoom=5,
                        mapbox_style="open-street-map")
fig.update_layout(mapbox_style="open-street-map")
fig.update_layout(margin={"r":0,"t":0,"l":0,"b":0})
fig.show()
In [37]:
df_count= pd.DataFrame({'Location':df.Location.value_counts().index,
                        'Country':df.Country.value_counts().values,
                        'Count':df.Country.value_counts().values})
In [38]:
#Calculate the percentage
df_count["Percentage"]=df_count['Count']/10
df_count
Out[38]:
Location Country Count Percentage
0 USA 213 213 21.3
1 China 108 108 10.8
2 United Kingdom 62 62 6.2
3 France 58 58 5.8
4 Japan 56 56 5.6
... ... ... ... ...
56 Uruguay 1 1 0.1
57 Slovak Republic 1 1 0.1
58 Estonia 1 1 0.1
59 Iceland 1 1 0.1
60 Macau 1 1 0.1

61 rows × 4 columns

In [43]:
#Pie plot
labels = df_count.Location
values = df_count.Country

fig = px.pie(values=values,  
             names=labels,
             hole=0.3)

fig.update_traces(textinfo='none')  # Remove percentage labels

fig.update_layout(
    width=1000, height=500,
    title_text="Count by countries",
    title_x=0.45, 
    title_yanchor="middle"
)

fig.show()
In [44]:
#Pie plot
labels = df.Continent.value_counts().index
values = df.Continent.value_counts().values

fig = px.pie(values=values,  
             names=labels,
             hole=0.3)


fig.update_layout(
    width=1000, height=500,
    title_text="Count By continent",
    title_x=0.50, 
    title_yanchor="middle"
)

fig.show()